Skip to main content

Scheduling

You never have to set a due date. Every time you grade a card, the app works out when you are likely to be on the verge of forgetting it and schedules the next review for then. This page explains how that works and which settings change it.

FSRS​

Audio Flashcards uses FSRS (Free Spaced Repetition Scheduler), a modern, well-tested scheduling algorithm. From a card's full history of grades it estimates two things:

  • Stability — how long the memory is likely to last.
  • Difficulty — how hard this particular card is for you.

The next due date follows from those. Cards you know drift further and further apart; cards you keep missing come back quickly until they stick.

A card is due once its due date has passed. New cards that have never been reviewed are due straight away.

What the grades do​

GradeEffect
AgainThe memory has lapsed. The card comes back soon — typically within minutes — and its stability drops.
HardRecalled with effort. A shorter interval than Good.
GoodRecalled normally. The standard interval, growing with each successful review.
EasyRecalled effortlessly. A longer interval than Good.

You do not have to guess: the time printed under each grade in the player is exactly when the card would come back. Early on those times are minutes and days; a card you have known for months shows months.

The history is the card's own, so the same grade gives different intervals on different cards.

Passive scheduling​

FSRS only learns from grades. That is a problem for listening sessions: if autoplay plays a card and you never grade it, FSRS does not know you heard it, the card stays due, and a looping session plays the same cards over and over.

The Passive Scheduler fixes that. Turn it on for a session (Edit Session > Passive Scheduler > Use Passive Scheduler) and every card that session plays without a grade — an Auto pass — pushes that card back along a curve of its own.

How it behaves:

  • It never overrides a grade. A card's due date is whichever is later: the FSRS date from your grades, or the passive date from your listens.
  • Only listens since your last grade count. Grading a card resets its passive progress.
  • Close listens count once. Auto passes less than 12 hours apart count as a single exposure, so a looping session does not race a card to the end of the curve in an afternoon.
  • It is per session. Each session has its own curve, so the same card can be due in your graded Daily Review and still resting in a background listening session.
  • It needs Record Review Progress. A practice session saves nothing, including listens.

Shape the curve​

The Curve graph plots how long the session waits after the 1st, 2nd, 3rd… passive exposure. Drag its three control points to change how quickly the delays grow. The chips under the graph read out the result — with the recommended curve, about 30h after the 1st listen, 2mo after the 10th, and 6mo from the 20th onward.

  • Graph Max Review and Graph Max Time set the edges of the graph, and so the furthest a control point can go.
  • Maturity Influence — Off, Light, Medium, or Strong — lets cards you have graded well wait longer between passive listens too. Off, the default, uses the curve alone.
  • Restore Recommended Curve puts everything back.

Settings that keep cards out of scheduling​

SettingWhereEffect
Record Review Progress offA sessionPractice mode. Grades and listens are not saved, and due dates and statistics do not change.
Due Cards Only offA sessionLets you review cards before they are due. Grading them still updates their schedule, unless the session is a practice session.
Deactivate FolderFolder SettingsThe folder's cards are left out of every session and statistic until you turn it back on.

To wipe one card's history and start it again as new, open the card and choose ⋯ > Card Statistics > Clear Card Statistics.

Spaced Repetition settings​

Open Review > Edit Review Session > Spaced Repetition on Apple or Android to adjust FSRS for that session:

  • Desired retention defaults to 90%. Choose 85%, 90%, or 95%, or a custom target between 80% and 97%. Higher targets shorten intervals and increase review work. Retention is an estimate, not a guarantee.
  • Maximum interval caps graded-review intervals at 1–36,500 days. Learning steps remain fixed at 1 and 10 minutes, with a 10-minute relearning step.
  • Save the review session to apply changes. Its schedules recalculate immediately from the existing review history; other sessions retain their own configuration. Canceling the session editor discards the draft. Its passive scheduler may postpone FSRS dates.
  • Restore Recommended Settings resets this session's draft to 90% and the default maximum interval. Review history is preserved.

Each session applies one FSRS configuration to all its cards. Moving or renaming cards and folders does not change their FSRS settings within a session. Review files contain grade timestamps; due dates and statistics are calculated with the selected session's current parameters. Recall charts compare observed recall with its current retention target.

Existing global defaults are copied into saved sessions once. Old folder overrides are ignored; existing sessions retain their session values. After migration, directory.meta and the retired global FSRS key do not control scheduling. New sessions start with recommended settings; duplicating a session copies its FSRS configuration.

Settings are saved with the other review-session settings in @reviewSessionPresets, using the selected local/iCloud storage on Apple and local storage on Android. Library backups include each session's FSRS configuration. Parameter optimization is not available in this version; FSRS uses its standard weights.